344 RADIATION UIOLOGY 



iiiactivation for very siricall doses (about 10 crj^.s X mm-' for 2537 A), 

 soon clian^iinjz; to a lotj;aritlimic rate as the dose increases (Beiizer cl al., 

 1950). The reason for this heluvvior is obscure; it does not seem to be due 

 to the presence of agf!;re}j;ates of \ irus particles. Other phages (e.g., Tl, 

 T7) show an initial logarithmic rate with a br(!ak to a slower rate for sur- 

 vivals lower than lO^-. The more resistant fraction of virus is not geneti- 

 cally dilTerent; it is possible that it is combined with screening materials. 

 Complications of this kind make calculations of iiiactivation rates and of 

 (juantum yields somewhat (luestionablc. 



Action spectra have been reported for several viruses (liivers and ( lates, 

 1028; Sturm et a/., 1932; dates, 1934; HoUaender and Duggar, 1936; 

 IloUaender and Oliphant, 1944). Comparisons were generally made only 

 for incident energies, however, by plotting the inverse of the incident dose 

 re(iuired to produce a constant amount of inactivation versus the wave 

 length. For most viruses the graph resembles the absorption curve of 

 nucleic acids with a minimum at 2400 A, a maximum around 2000 A, and 

 very low effectiveness beyond 3000 A. For some \'iruses, howe\'er, the 

 maximum and minimum at 2()00 and 2400 A, respectively, are much le.ss 

 pronounced than for other viru.ses. There is no clear correlation between 

 total luicleic acid content and type of action spectrum since vaccinia virus 

 and tobacco mosaic virus have approximately the same nucleic acid con- 

 tent (in percentage of dry weight), yet give different action spectra. It 

 has been pointed out that the viruses, whose action spectra are less similar 

 to the absorption spectrum of nucleic acid, siipposedly contain the ribose 

 instead of the deoxyribose type (HoUaender, 1946). 



Although these results indicate that, at least in most cases, a large pro- 

 portion of the effective radiation is absorbed by the nucleic acid of the 

 \irus particles, they do not indicate the relative effectiveness of quanta 

 absorbed by different virus components. If the part of a radiation 

 absorbed by luicleic acid and that absorbed by other components, e.g., 

 proteins, were equally effective in producing inactivation, the greater 

 absorption coefficient of the nucleic acids for most ultraviolet wave lengths 

 would cause them to appear as the main contributors to the effective 

 absorption whenever they are present in the amounts found in many 

 viruses (5-40 per cent of the dry weight). 



More information could be gained from action-spectrum studies based 

 on measurements not of incident radiation energy but of actual (juantum 

 yield. Absorption measurements on purified virus preparations are easily 

 feasible, yet surprisingly few data on quantum yield for virus inactivation 

 have been reported. In most cases they are for one wa\(> length only, the 

 2537 A line of mercury. One difficulty, of course, is that the actual virus 

 content of a preparation, in terms of particles per milliliter, is .seldom 

 accurately known. For tobacco mosaic virus and 2537 .\ the values of 

 2.6 X 10-^ (Uber, 1941) and 4.3 X 10-'^ (Oster and McLaren. 1950) have 

 been reported for the cinantum yield, the latter \alue being probably more 



